441-447 21/7/06 17:25 Page 441

INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 18: 441-447, 2006 441

Heterogeneous expression of DNA-dependent kinase in esophageal cancer and normal epithelium

NORIO TONOTSUKA1,2, YOSHIO HOSOI4, SHUKICHI MIYAZAKI2, GO MIYATA2, KO SUGAWARA2, TAKAHIRO MORI3, NORIAKI OUCHI3, SUSUMU SATOMI2, YOSHIHISA MATSUMOTO4, KEIICHI NAKAGAWA5, KIYOSHI MIYAGAWA4 and TETSUYA ONO1

1Department of Radiation Research, 2Division of Advanced Surgical Science and Technology and 3Division of Surgical Oncology, Faculty of Medicine, Tohoku University, Sendai 980-8575; Departments of 4Radiation Research and 5Radiology, Faculty of Medicine, University of Tokyo, Tokyo 113-0033, Japan

Received February 10, 2006; Accepted April 13, 2006

Abstract. Esophageal cancer tissues and adjacent normal of the catalytic subunit of DNA-PK (DNA-PKcs) and a mucosae in 13 patients with primary esophageal cancer were heterodimer of and (1). DNA-PK plays a crucial examined for quantitative differences in DNA-dependent role in the repair of DNA double-strand breaks (DSBs) induced protein kinase (DNA-PK) activity and for expressions of by ionizing radiation and chemotherapeutic agents (1). Cells Ku70, Ku80 and DNA-PKcs by Western blotting and lacking DNA-PK activity because of defects in the DNA-PK immunohistochemistry. The tumor tissues showed higher components show hypersensitivity to ionizing radiation and DNA-PK activity than the normal mucosae. Protein levels of chemotherapeutic agents (2-6). Ku70, Ku80 and DNA-PKcs correlated with DNA-PK There have been many reports on the examination whether activities in the tumor tissues. Immunohistochemical analysis DNA-PK activity correlates with radiation sensitivity and revealed that Ku70, Ku80 and DNA-PKcs located pre- whether it can be a parameter indicating the sensitivity to dominantly in the nuclei in both the tumor tissues and radiotherapy and/or chemotherapy. Suppression of DNA-PK normal mucosae. In the normal epithelium, Ku70, Ku80 and activity by a phosphatidylinositol 3-kinase inhibitor wort- DNA-PKcs were expressed only in the nuclei of the basal mannin, antisense Ku70/DNA-PKcs, or small inhibitory RNA cell layers and not in those of the lumenal cell layers. In the for DNA-PKcs sensitized cells to ionizing radiation (7-10). tumor tissues, the expressions of DNA-PK proteins showed Polischouk et al reported that levels of DNA-PK activity intratumoral heterogeneity. The different portions in the same associate with the proficiency in rejoining of DNA double- tumor showed different expression levels of DNA-PK proteins, strand breaks (11). These results indicate that levels of DNA- and even each tumor cell showed different expression levels. PK activity correlates with cellular sensitivity to ionizing These results suggest that cell differentiation and tumor radiation in the cells under the same genetic background. progression affect cellular DNA-PK protein levels and its Relationship between radiation sensitivity and DNA-PK activity. Furthermore, the intratumoral heterogeneity of activity/protein levels under different genetic backgrounds DNA-PK protein expression in esophageal cancer cells/ has been investigated in cultured cell lines or tissue specimens tissues also suggests the difficulty in prediction of radio- or obtained from patients with cancer (11-20). The results chemo-sensitivity of the tumor through estimation of DNA-PK reported were contradictory and it is still unclear whether activity/protein levels in tumor specimens. DNA-PK can be a parameter indicating radiation sensitivity under different genetic backgrounds. Introduction To predict curability of the tumor after radiotherapy or chemotherapy, the sensitivities in both tumor tissue and DNA-dependent protein kinase (DNA-PK) is a nuclear adjacent normal tissue should be assessed. We previously protein with serine/threonine kinase activity and it composed examined the DNA-PK activities in tumor tissues and adjacent normal tissues in patient with colorectal cancer and found ______that the tumor tissues showed higher DNA-PK activity than the adjacent normal tissues in 11 out of 12 patients, which Correspondence to: Dr Yoshio Hosoi, Department of Radiation suggests poor curability of the tumors after radiation therapy Research, Faculty of Medicine, University of Tokyo, 7-3-1 Hongo, alone under the condition that the DNA-PK activity correlates Bunkyoku, Tokyo 113-0033, Japan with radiation sensitivity (21). In the present study, we E-mail: [email protected] examined DNA-PK activities and protein levels of the tumor tissues and the adjacent normal mucosae in patients with Key words: DNA-dependent protein kinase, esophageal cancer, esophageal cancer by the standard kinase activity assay, Ku70, Ku80, DNA-PKcs Western blotting and immunohistochemistry. The results revealed that DNA-PK activity was higher in esophageal 441-447 21/7/06 17:25 Page 442

442 TONOTSUKA et al: HETEROGENEIC EXPRESSION OF DNA-PK IN ESOPHAGEAL CANCER

Table I. Characteristics of the patients. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Patient No. Sex Positiona Histologyb Differentiation pTNMc ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 1 M Lt SCC Poorly differentiated III 2 F Mt SCC Well differentiated III 3 F Mt SCC Poorly differentiated I 4 M Ce SCC Poorly differentiated III 5 M Mt SCC Moderately differentiated I 6 M Mt SCC Well differentiated III 7 F Lt Undifferentiated Undifferentiated IIB 8 F Lt SCC Moderately differentiated IIB 9 M Mt Basaloid - III 10 F Mt SCC Well differentiated I 11 M Lt SCC Poorly differentiated IIA 12 M Ae SCC Poorly differentiated IIB 13 M Mt SCC Poorly differentiated I ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– aCe, cervical esophagus; Mt, middle thoracic esophagus; Lt, lower thoracic esophagus; Ae, abdominal esophagus. bSCC, squamous cell carcinoma; undiffrentiated, Undifferentiated carcinoma (small cell type); basaloid, basaloidcarcinoma. cpTNM, pathological stage grouping according to TNM classification. –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

cancer tissues than in the normal mucosae and that the levels DNA-PK activity. DNA-PK activity was assayed as previously of DNA-PK proteins correlated with the DNA-PK activity. described, with a synthetic peptide (EPPLSQEAFAD Expressions of DNA-PK proteins in tumor tissues were found LWKK) (7). The whole-cell or tissue extracts were incubated to be heterogeneous, which suggests the difficulty in prediction in 20 μl of kinase buffer [20 mM HEPES-NaOH (pH 7.2),

of radio- or chemo-sensitivity of the tumors thorough the 100 mM KCl, 5 mM MgCl2, 1 mM DTT, 0.5 mM NaF, 0.5 mM examination of DNA-PK activity/protein levels in tumor ß-glycerophosphate, 0.2 mM ATP, 10 μCi/ml [Á-32P]ATP in specimens of esophageal cancer. the presence of 0.01 mg/ml sonicated salmon sperm DNA and 0.5 mg/ml substrate peptide] at 37˚C for 15 min. The final Materials and methods protein concentration in the reaction mixture was 37.5 μg/ml. The reactions were stopped by the addition of 20 μl of 30% Tissue specimens. All esophageal tumors and adjacent normal acetic acid and the mixtures were spotted onto P81 paper disks tissues were obtained at the time of surgery at Tohoku (Whatman International Ltd., Maidstone, UK). The disks University Hospital from 1999 to 2000. Informed consent was were washed 4 times in 15% acetic acid. Radioactivity in the received from all patients. The patients received neither radio- paper disks was measured in a liquid scintillation counter. therapy nor chemotherapy before surgery. Characteristics of the patients are shown in Table I. Western blotting. Whole cell extracts or tissue extracts were lysed in the electrophoresis sample buffer [62.5 mM Tris Cells. LM217 is an SV40 transformed human fibroblast cell (pH 6.8), 2% SDS, 5% glycerol, 0.003% bromophenol blue, line derived from HS27 (22). LM217 was used as control in 1% ß-mercaptoethanol] and boiled for 5 min. The lysate was measurement of DNA-PK activity and Western blotting. resolved by electrophoresis using a gradient gel (Daiichi Pure Chemicals Co., Ltd., Tokyo, Japan), and was electrophore- Whole-cell and tissue extracts. Whole-cell extracts and tissue tically transferred to polyvinylidene difluoride membranes extracts were prepared by a modification of the methods of (Bio-Rad). The membranes were then proved with anti-Ku70 Finnie et al and Dignam et al (23,24). The samples were antibody, anti-Ku80 antibody, anti-DNA-PKcs antibody Ab-4 washed twice with Tris-buffered saline [2 mM Tris (pH 7.2), (Cocktail) (NeoMarkers, Fremont, CA) or Anti-GAPDH anti- 150 mM NaCl], homogenized using a hand-operated homo- body (Trevigen, Inc., Gaithersburg, MD). The anti-Ku70 and genizer (Eppendorf, Hamburg, Germany), then suspended in anti-Ku80 antibodies used were raised in our laboratory as 100 μl of a low-salt buffer [10 mM HEPES (pH 7.2), 25 mM previously reported (19). The antigen-antibody complexes KCl, 10 mM NaCl, 1.1 mM MgCl2, 1 mM EDTA, 1 mM were detected by the ECL Plus™ Western blotting detection EGTA, 1 mM PMSF, 1 mM DTT, 1 μg/ml pepstatin, 1 μg/ml reagents (Amersham Pharmacia Biotech Inc., Piscataway, NJ), leupeptin, 1 μg/ml antipain], and then frozen in liquid nitrogen with horseradish peroxidase-conjugated antibodies. The images and thawed at 30˚C three times. After a 60-min incubation at were analyzed with the Scion Image Beta 4.02 Win software 4˚C, the suspension was adjusted to 0.4 M KCl by adding (Scion Corporation, Frederick, MD) to quantify the densities of 3.5 M KCl, incubated for 30 min at 4˚C, and centrifuged for bands corresponding to Ku70, Ku80, DNA-PKcs and GAPDH. 10 min at 15,000 rpm. The supernatant was designated as the whole-cell extract (25). Protein concentrations were determined Immunohistochemistry for DNA-PK proteins. Formalin-fixed with the Bio-Rad protein assay (Bio-Rad, Hercules, CA). and paraffin-embedded tissue specimens were deparaffinized, 441-447 21/7/06 17:25 Page 443

INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 18: 441-447, 2006 443

Figure 1. DNA-PK activities of tumor tissues and adjacent normal mucosae in 13 patients with esophageal cancer. DNA-PK activities are expressed as values relative to that of LM217, which is set to a value of 1. (A) Mean ± SD of DNA-PK activities of three different specimens of tumor tissues and normal tissues in each patient, *P<0.05. (B) Average values of DNA-PK activities of tumor tissues and normal tissues in 13 patients. Mean ± SD, *P<0.05.

cut to 2-μ sections, and stained by the labeled streptavidin biotin (LSAB) technique. Briefly, the sections were incubated with 3% hydrogen peroxide in methanol, and then incubated with 1% bovine serum for 30 min for blocking. The sections were incubated with primary antibody overnight at room temperature, and incubated with biotinylated secondary anti- body (Nichirei Biosciences Inc., Tokyo, Japan) for 30 min at room temperature. Then, streptavidin/biotin complex was applied for 30 min (Nichirei Biosciences Inc.), followed by a 30 min incubation in 3,3'-diaminobnzidine substrate, yielding a brown reaction product. Sections were counterstained with hematoxylin and mounted under a coverslip. Figure 2. Expression of DNA-PK proteins in normal mucosae and tumor tissues. (A) Western blot analysis for Ku70, Ku80, DNA-PKcs and GAPDH Evaluation. The intensity of staining in immunohistochemistry in the extracts prepared from tumor tissues (T) and normal mucosae (N) was evaluated by the expression score reported by Rigas et al obtained from patients No. 2, 4, 6, 7 and 12. (B-D) Relative protein levels of (26). The intensity of staining was rated according to the Ku70, Ku80 and DNA-PKcs in the 5 patients. Histograms represent the following scale: 2 = intense brown staining, 1 = light brown scanning densitometric analysis of Western blots. Each value is normalized for GAPDH and expressed as a value relative to that of LM217, which is set staining, 0 = no staining. In each sample, the percentage of to a value of 1. Mean ± SD of the data from three different tissue extracts cells expressing each protein was determined. To obtain a used in the measurement of DNA-PK activity. (E) Average values of Ku70, numerical assessment of the expression of each protein, we Ku80 and DNA-PKcs protein levels of tumor tissues and normal tissues in calculated the multiple of the intensity of staining by the the 5 patients. Mean ± SD, *P<0.05. percentage of cells expressing a protein for each sample. In tumor tissues, we assessed the expression score at the growing edge of the carcinoma, which is defined as the invasive tip, Results because tumor growth largely depends on the proliferative kinetics at the invasive tip (27). In normal tissues, the DNA-PK activity of tumor tissues and normal mucosae. We expression score was assessed at the normal mucosae. Each examined DNA-PK activities of tumor tissues and adjacent expression score was expressed as mean ± SD of three scores normal mucosae in 13 patients with esophageal cancer. In 5 evaluated at the three different fields in the invasive tip or in patients, DNA-PK activity was significantly higher in the normal mucosae respectively with original magnification of tumor tissue compared with the normal mucosa (Fig. 1A). x400. Average value of the DNA-PK activities of the tumor tissues in the 13 patients was significantly higher than that of the Statistics. Homogeneity of variance was tested by the F-test. normal mucosae (Fig. 1B). When variance was homogeneous, Student's t-test was used. When variance was heterogeneous, Welch's test was used. Expression of DNA-PK proteins in tumor tissues and normal For analysis of correlation coefficient, the distributions of mucosae. Next, we examined DNA-PK protein levels by variables were tested first. When variables were normally Western blotting in order to investigate whether the variety distributed, Peason's correlation coefficient test was used. of DNA-PK activities observed in tumor tissues and normal When variables were not normally distributed, Spearman's mucosae depended on the DNA-PK protein levels. For the rank correlation coefficient test was used. Differences were Western blotting, we used the same tissue extracts as used in considered to be statistically significant at P<0.05. the assessments of DNA-PK activity. Western blotting was 441-447 21/7/06 17:25 Page 444

444 TONOTSUKA et al: HETEROGENEIC EXPRESSION OF DNA-PK IN ESOPHAGEAL CANCER

conducted using the tissue extracts from the 5 patients Table II. Correlation coefficient. whose had enough tissue extracts for the Western blot ––––––––––––––––––––––––––––––––––––––––––––––––– analysis (Fig. 2A). In 2 out of the 5 patients, levels of Ku70, A, Correlation coefficients in normal tissues Ku80 and DNA-PKcs proteins were significantly higher in the ––––––––––––––––––––––––––––––––––––––––––––––––– tumor tissue compared with the normal mucosae (Fig. 2B-D). Ku70 Ku80 DNA-PKcs Average value of Ku70 protein levels in the 5 patients was –––––––––––– ––––––––––––– –––––––––––– P CC P CC P CC significantly higher in the tumor tissue (Fig. 2E). Significant ––––––––––––––––––––––––––––––––––––––––––––––––– correlation was observed between DNA-PK activity and DNA-PK 0.323 -0.491 0.258 -0.550 0.975 0.975 protein levels of Ku70, Ku80 and DNA-PKcs in tumor tissues activity of the 5 patients (Table IIB). Ku70 and Ku80 protein levels correlated with DNA-PKcs protein level in the tumor tissues Ku70 - - 0.0401 0.816 0.725 0.157 (Table IIB). Ku70 protein level correlated with Ku80 protein level in the normal tissues (Table IIA). Ku80 0.041 0.816 - - 0.544 0.271 DNA-PKcs 0.725 0.157 0.544 0.271 - - Immunohistochemical analysis in normal mucosae. Esophageal ––––––––––––––––––––––––––––––––––––––––––––––––– epithelium consists of non-keratinized stratified squamous B, Correlation coefficients in tumor tissues cells as shown in Fig. 3A. Stainings for Ku70, Ku80 and ––––––––––––––––––––––––––––––––––––––––––––––––– DNA-PKcs were predominantly nuclear and they showed a Ku70 Ku80 DNA-PKcs similar pattern (Fig. 3B-E). In the epithelium, Ku70, Ku80 ––––––––––––– ––––––––––––– ––––––––––––– and DNA-PKcs were expressed exclusively in the nuclei of the P CC P CC P CC basal cell layers and not in those of the lumenal cell layers ––––––––––––––––––––––––––––––––––––––––––––––––– (Fig. 3B-D). The DNA-PK proteins were expressed in almost DNA-PK 0.0476 0.832 0.00580 0.937 0.0100 0.917 all the nuclei in the middle-basal cell layers of the epithelium, activity whereas they were not expressed in some nuclei in the most Ku70 - - 0.101 0.728 0.000285 0.986 basal cell layers (Fig. 3B-D). In lamina propria and muscularis mucosa, the DNA-PK proteins were expressed in about half Ku80 0.101 0.728 - - 0.000284 0.986 of the nuclei (Fig. 3B-D). DNA-PKcs 0.000285 0.986 0.000284 0.986 - - ––––––––––––––––––––––––––––––––––––––––––––––––– Immunohistochemical analysis in tumor tissues. In tumor CC, correlation coefficient. Bold values indicating P<0.05. tissues, stainings for Ku70, Ku80 and DNA-PKcs were also ––––––––––––––––––––––––––––––––––––––––––––––––– predominantly nuclear and they showed a similar pattern in each tumor (Figs. 4 and 5). The intensity of staining was heterogeneous in the tumor tissues (Figs. 4B-D and 5B-D). recombination (HR) are the two major repair mechanisms for The different portions in the same tumor showed different DSBs, and NHEJ plays the most important role in expression levels of DNA-PK proteins, and even each tumor mammalian cells (28). Radiation sensitivity can be possibly cell showed different expression levels (Figs. 4F-H and 5F-H). predicted through the quantitative evaluation of DNA-PK as a These heterogeneic staining patterns of the DNA-PK proteins key enzyme for NHEJ because DNA-PK activity has relevance could be observed in all the tumors examined. to cellular sensitivity to ionizing radiation and DNA-PK protein levels correlate with DNA-PK activity (7-11). In the present A semi-quantitative assessment of DNA-PK proteins. To study, the immuno-histochemical analysis revealed the assess the expression of DNA-PK proteins semi-qantitatively intratumoral heterogeneity of DNA-PK protein expression in immunohistochemical examination, the intensity of (Figs. 4 and 5). The different portions in the same tumor staining was evaluated by the expression score described in showed different expression levels of DNA-PK proteins, and Materials and methods. The expression score was significantly even each tumor cell showed different expression levels (Figs. higher in the tumor tissue than in the normal mucosa in 7 4 and 5). This heterogeneous expression of DNA-PK proteins patients for Ku70, 4 patients for Ku80 and 6 patients for in esophageal cancer tissues suggests the difficulty in DNA-PKcs (Fig. 6A-C). In 2 patients, the expression score prediction of curability of the tumors after radiotherapy or for DNA-PKcs was lower in the tumor tissue than in the chemotherapy through evaluation of DNA-PK protein levels normal mucosa (Fig. 6C). The average values of expression in the tumor specimens because the curability will reflect the scores for Ku70 and Ku80 in 13 patients were significantly DNA-PK protein levels in the cells that express DNA-PK higher in the tumor tissue than in the normal mucosa (Fig. 6D). most abundantly in the tumor under the condition that radiation sensitivity correlates with DNA-PK activity. Discussion In the present study, DNA-PK activity correlated with protein levels of Ku70, Ku80 and DNA-PKcs, and protein Prediction of radio- and chemo-sensitivity of normal and tumor levels of Ku70 and Ku80 correlated with the levels of DNA- tissues before the treatment will provide crucial information PKcs in tumor tissues (Table II). These results correspond with to find the best treatment method for each patient with previous reports (16,18,20,21). regions of Ku70, cancer. In most of the cells, cell survival after X-irradiation Ku80 and DNA-PKcs contain consensus Sp1 recognition depends on the yield of DNA DSBs and the repair of them. elements and therefore these are supposed to be regulated Non-homologous end-joining (NHEJ) and homologous by the same transcriptional factor, Sp1 (21,29,30). 441-447 21/7/06 17:25 Page 445

INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 18: 441-447, 2006 445

Figure 3. The expression of DNA-PK proteins in esophageal mucosae from patient No. 7. (A) Staining with hematoxylin and eosin. (B) Immunostaining for Ku70. (C) Immunostaining for Ku80. (D) Immunostaining for DNA-PKcs. Original magnification x100.

Figure 5. The expression of DNA-PK proteins in esophageal cancer tissues from patient No. 13. (A and E) Staining with hematoxylin and eosin. (B and F) Immunostaining for Ku70. (C and G) Immunostaining for Ku80. (D and H) Immunostaining for DNA-PKcs. Original magnification x16 (A-D) and x400 (E-H).

Sp1 biding sites are located in the promoter regions of a number of growth-regulated genes including insulin-like growth factor-binding protein 2 and vascular/endothelial growth factor (31,32). Suppression of Sp1 by dominant negative Sp1 and Sp1-site-decoy oligonucleotides induces cell growth arrest (32,33). These results suggest involvement of Sp1 in growth-regulation. In the present study, DNA-PK proteins were highly expressed in the nuclei of the basal cell layers of the normal epithelium and were not expressed in luminal cell layers (Fig. 3), which may reflect the growth of the stem cells located in the basal cell layers and the growth arrest of the differentiated cells in the luminal cell layers. In the present study, the different portions in the same tumor showed different expression levels of DNA-PK proteins, and even each tumor cell showed different expression levels (Figs. 4 and 5). In esophageal cancer, intratumoral Figure 4. The expression of DNA-PK proteins in esophageal cancer tissues heterogeneity has been reported in expressions of many from patient No. 7. (A and E) Staining with hematoxylin and eosin. (B and F) proteins includeing transcription factor Ets-1, cyclin B1, Immunostaining for Ku70. (C and G) Immunostaining for Ku80. (D and H) Immunostaining for DNA-PKcs. Original magnification x16 (A-D) and cyclin D1, retinoblastoma protein, cyclooxygenase-2, GAGE, x400 (E-H). NY-ESO-1, MAGE-A, SSX, E-cadherin and ·-catein (34-40). 441-447 21/7/06 17:25 Page 446

446 TONOTSUKA et al: HETEROGENEIC EXPRESSION OF DNA-PK IN ESOPHAGEAL CANCER

3. Lees-Miller SP, Godbout R, Chan DW, Weinfeld M, Day RS III, Barron GM and Allalunis-Turner J: Absence of p350 subunit of DNA-activated protein kinase from a radiosensitive human cell line. Science 267: 1183-1185, 1995. 4. Gu Y, Jin S, Gao Y, Weaver DT and Alt FW: Ku70-deficient embryonic stem cells have increased ionizing radiosensitivity, defective DNA end-binding activity, and inability to support V(D)J recombination. Proc Natl Acad Sci USA 94: 8076-8081, 1997. 5. Nussenzweig A, Sokol K, Burgman P, Li L and Li GC: Hyper- sensitivity of Ku80-deficient cell lines and mice to DNA damage: the effects of ionizing radiation on growth, survival, and development. Proc Natl Acad Sci USA 94: 13588-13593, 1997. 6. Gao Y, Chaudhuri J, Zhu C, Davidson L, Weaver DT and Alt FW: A targeted DNA-PKcs-null mutation reveals DNA-PK- independent functions for in V(D)J recombination. Immunity 9: 367-376, 1998. 7. Hosoi Y, Miyachi H, Matsumoto Y, Ikehata H, Komura J, Ishii K, Zhao HJ, Yoshida M, Takai Y, Yamada S, Suzuki N and Ono T: A phosphatidylinositol 3-kinase inhibitor wortmannin induces radioresistant DNA synthesis and sensitizes cells to bleomycin and ionizing radiation. Int J Cancer 78: 642-647, 1998. 8. Omori S, Takiguchi Y, Suda A, Sugimoto T, Miyazawa H, Takiguchi Y, Tanabe N, Tatsumi K, Kimura H, Pardington PE, Chen F, Chen DJ and Kuriyama T: Suppression of a DNA double-strand break repair , Ku70, increases radio- and chemosensitivity in a human lung carcinoma cell line. DNA Repair 1: 299-310, 2002. 9. Sak A, Stuschke M, Wurm R, Schroeder G, Sinn B, Wolf G and Figure 6. The expression of DNA-PK proteins in normal mucosae and tumor Budach V: Selective inactivation of DNA-dependent protein tissues estimated by the expression score. The expression score is the kinase with antisense oligodeoxynucleotides: consequences for product of the intensity of expression multiplied by the percentage of cell the rejoining of radiation-induced DNA double-strand breaks expressing a given protein, and it assesses numerically the expression of and radiosensitivity of human cancer cell lines. Cancer Res 62: 6621-6624, 2002. each protein. Details are described in Materials and methods. (A and C) 10. Collis SJ, Swartz MJ, Nelson WG and De Weese TL: Enhanced Expression scores for Ku70, Ku80 and DNA-PKcs in patients No. 1-13, radiation and chemotherapy-mediated cell killing of human respectively. Mean ± SD of three expression scores. (D) Average values of cancer cells by small inhibitory RNA silencing of DNA repair the expression scores in the 13 patients. Mean ± SD, *P<0.05. factors. Cancer Res 63: 1550-1554, 2003. 11. Polischouk AG, Cedervall B, Ljungquist S, Flygare J, Hellgren D, Grenman R and Lewensohn R: DNA double-strand break repair, DNA-PK, and DNA ligases in two human squamous carcinoma cell lines with different radiosensitivity. Int J Radiat Oncol Biol It has been reported that esophageal cancer is genetically Phys 43: 191-198, 1999. 12. Allalunis-Turner MJ, Lintott LG, Barron GM, Day RS and heterogeneous and it consists of various sub-clones with Lees-Miller SP: Lack of correlation between DNA-dependent different status of p53 gene as a result of tumor progression protein kinase activity and tumor cell radiosensitivity. Cancer (41-43). This genetic heterogeneity in esophageal cancer may Res 55: 5200-5202, 1995. 13. Kasten U, Plottner N, Johansen J, Overgaard J and Dikomey E: underlie the heterogeneous expression of the proteins Ku70/80 and DNA-dependent protein kinase mentioned above and that of DNA-PK proteins as reported in (DNA-PK) activity do not correlate with double-strand break this study. Furthermore, the contribution of increased Sp1 (dsb) repair capacity and cellular radiosensitivity in normal human fibroblasts. Br J Cancer 79: 1037-1041, 1999. activity to tumor progression has been reported (33). 14. Sirzen F, Nilsson A, Zhivotovsky B and Lewensohn R: DNA- Upregulation of Sp1 in the process of tumor progression may dependent protein kinase content and activity in lung carcinoma increase the DNA-PK activity/protein levels of the tumor cells, cell lines: correlation with intrinsic radiosensitivity. Eur J Cancer 35: 111-116, 1999. as shown in Figs. 1, 2 and 6, and cause the acquisition of radio- 15. Vaganay-Juery S, Muller C, Marangoni E, Abdulkarim B, resistant phenotype. Deutsch E, Lambin P, Calsou P, Eschwege F, Salles B, Joiner M In summary, DNA-PK activity was higher in the esophageal and Bourhis J: Decreased DNA-PK activity in human cancer cells exhibiting hypersensitivity to low-dose irradiation. Br J tumor tissues compared with the adjacent normal mucosae, Cancer 83: 514-518, 2000. and DNA-PK protein levels correlated with DNA-PK activity. 16. Bjork-Eriksson T, West C, Nilsson A, Magnusson B, Svensson M, Expression of DNA-PK proteins showed intratumoral hetero- Karlsson E, Slevin N, Lewensohn R and Mercke C: The immuno- histochemical expression of DNA-PKCS and Ku (p70/p80) in geneity in esophageal cancer tissues, making difficult the head and neck cancers: relationships with radiosensitivity. Int J prediction of curability of the tumors after radiotherapy or Radiat Oncol Biol Phys 45: 1005-1010, 1999. chemotherapy by estimation of DNA-PK activity/protein levels 17. Zhao HJ, Hosoi Y, Miyachi H, Ishii K, Yoshida M, Nemoto K, Takai Y, Yamada S, Suzuki N and Ono T: DNA-dependent in tumor specimens. protein kinase activity correlates with Ku70 expression and radiation sensitivity in esophageal cancer cell lines. Clin Cancer References Res 6: 1073-1078, 2000. 18. Wilson CR, Davidson SE, Margison GP, Jackson SP, Hendry JH and West CM: Expression of Ku70 correlates with survival in 1. Gottlieb TM and Jackson SP: The DNA-dependent protein carcinoma of the cervix. Br J Cancer 83: 1702-1706, 2000. kinase: requirement for DNA ends and association with Ku 19. Sakata K, Matsumoto Y, Tauchi H, Satoh M, Oouchi A, antigen. Cell 72: 131-142, 1993. Nagakura H, Koito K, Hosoi Y, Suzuki N, Komatsu K and 2. Jeggo PA, Taccioli GE and Jackson SP: Menage a trois: double Hareyama M: Expression of genes involved in repair of DNA strand break repair, V(D)J recombination and DNA-PK. double-strand breaks in normal and tumor tissues. Int J Radiat Bioessays 17: 949-957, 1995. Oncol Biol Phys 49: 161-167, 2001. 441-447 21/7/06 17:25 Page 447

INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 18: 441-447, 2006 447

20. Komuro Y, Watanabe T, Hosoi Y, Matsumoto Y, Nakagawa K, 33. Ishibashi H, Nakagawa K, Onimaru M, Castellanous EJ, Tsuno N, Kazama S, Kitayama J, Suzuki N and Nagawa H: The Kaneda Y, Nakashima Y, Shirasuna K and Sueishi K: Sp1 expression pattern of Ku correlates with tumor radiosensitivity decoy transfected to carcinoma cells suppresses the expression and disease free survival in patients with rectal carcinoma. Cancer of vascular endothelial growth factor, transforming growth 95: 1199-1205, 2002. factor beta1, and tissue factor and also cell growth and invasion 21. Hosoi Y, Watanabe T, Nakagawa K, Matsumoto Y, Enomoto A, activities. Cancer Res 60: 6531-6536, 2000. Morita A, Nagawa H and Suzuki N: Up-regulation of DNA- 34. Saeki H, Kuwano H, Kawaguchi H, Ohno S and Sugimachi K: dependent protein kinase activity and Sp1 in colorectal cancer. Expression of ets-1 transcription factor is correlated with Int J Oncol 25: 461-468, 2004. penetrating tumor progression in patients with squamous cell 22. Murnane JP: Inducible gene expression by DNA rearrangements carcinoma of the esophagus. Cancer 89: 1670-1676, 2000. in human cells. Mol Cell Biol 6: 549-558, 1986. 35. Murakami H, Furihata M, Ohtsuki Y and Ogoshi S: Determination 23. Finnie NJ, Gottlieb TM, Blunt T, Jeggo PA and Jackson SP: of the prognostic significance of cyclin B1 overexpression in DNA-dependent protein kinase activity is absent in xrs-6 cells: patients with esophageal squamous cell carcinoma. Virchows implications for site-specific recombination and DNA double- Arch 434: 153-158, 1999. strand break repair. Proc Natl Acad Sci USA 92: 320-324, 36. Ishikawa T, Furihata M, Ohtsuki Y, Murakami H, Inoue A and 1995. Ogoshi S: Cyclin D1 overexpression related to retinoblastoma 24. Dignam JD, Lebovitz RM and Roeder RG: Accurate transcription protein expression as a prognostic marker in human oesophageal initiation by RNA polymerase II in a soluble extract from squamous cell carcinoma. Br J Cancer 77: 92-97, 1998. isolated mammalian nuclei. Nucleic Acids Res 11: 1475-1489, 37. Xing EP, Yang GY, Wang LD, Shi ST and Yang CS: Loss of 1983. heterozygosity of the Rb gene correlates with pRb protein 25. Hosoi Y, Kawamura M, Ido T, Takai Y, Ishii K, Nemoto K, expression and associates with p53 alteration in human Ono T, Kimura S and Sakamoto K: Sensitization of cells to esophageal cancer. Clin Cancer Res 5: 1231-1240, 1999. ionizing radiation by chlorin e6Na. Radiat Oncol Investig 6: 38. Abdalla SI, Sanderson IR and Fitzgerald RC: Effect of 151-156, 1998. inflammation on cyclooxygenase (COX)-2 expression in benign 26. Rigas B, Borgo S, Elhosseiny A, Balatsos V, Manika Z, Shinya H, and malignant oesophageal cells. 26: 1627- Kurihara N, Go M and Lipkin M: Decreased expression of 1633, 2005. DNA-dependent protein kinase, a DNA repair protein, during 39. Akcakanat A, Kanda T, Tanabe T, Komukai S, Yajima K, human colon carcinogenesis. Cancer Res 61: 8381-8384, 2001. Nakagawa S, Ohashi M and Hatakeyama K: Heterogeneous 27. Chino O, Makuuchi H, Shimada H, Machimura T, Mitomi T expression of GAGE, NY-ESO-1, MAGE-A and SSX proteins and Osamura RY: Assessment of the proliferative activity of in esophageal cancer: Implications for immunotherapy. Int J superficial esophageal carcinoma using MIB-1 immunostaining Cancer 118: 123-128, 2006. for the Ki-67 antigen. J Surg Oncol 67: 18-24, 1998. 40. Kadowaki T, Shiozaki H, Inoue M, Tamura S, Oka H, Doki Y, 28. Yoshida M, Hosoi Y, Miyachi H, Ishii N, Matsumoto Y, Iihara K, Matsui S, Iwazawa T and Nagafuchi A: E-cadherin Enomoto A, Nakagawa K, Yamada S, Suzuki N and Ono T: and alpha-catenin expression in human esophageal cancer. Roles of DNA-dependent protein kinase and ATM in cell-cycle- Cancer Res 54: 291-296, 1994. dependent radiation sensitivity in human cells. Int J Radiat Biol 41. Kaketani K, Saito T, Kuwahara A, Shimoda K, Miyahara M, 78: 503-512, 2002. Chikuba K, Etoh K and Kobayashi M: DNA stem line 29. Ludwig DL, Chen F, Peterson SR, Nussenzweig A, Li GC and heterogeneity in esophageal cancer accurately identified by flow Chen DJ: Ku80 gene expression is Sp1-dependent and sensitive to cytometric analysis. Cancer 72: 3564-3570, 1993. CpG methylation within a novel cis element. Gene 199: 181-194, 42. Kuwabara S, Ajioka Y, Watanabe H, Hitomi J, Nishikura K and 1997. Hatakeyama K: Heterogeneity of p53 mutational status in 30. Fujimoto M, Matsumoto N, Tsujita T, Tomita H, Kondo S, esophageal squamous cell carcinoma. Jpn J Cancer Res 89: Miyake N, Nakano M and Niikawa N: Characterization of the 405-410, 1998. promoter region, first ten exons and nine intron-exon boundaries 43. Shi ST, Yang GY, Wang LD, Xue Z, Feng B, Ding W, Xing EP of the DNA-dependent protein kinase catalytic subunit gene, and Yang CS: Role of p53 gene mutations in human esophageal DNA-PKcs (XRCC7). DNA Res 4: 151-154, 1997. carcinogenesis: results from immunohistochemical and mutation 31. Milanini J, Vinals F, Pouyssegur J and Pages G: p42/p44 MAP analyses of carcinomas and nearby non-cancerous lesions. kinase module plays a key role in the transcriptional regulation Carcinogenesis 20: 591-597, 1999. of the vascular endothelial growth factor gene in fibroblasts. J Biol Chem 273: 18165-18172, 1998. 32. Kutoh E, Margot JB and Schwander J: Identification and characterization of the putative retinoblastoma control element of the rat insulin-like growth factor binding protein-2 gene. Cancer Lett 136: 187-194, 1999.